[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-post-en-/how-many-languages-do-you-support":3},{"page":4,"surround":514},{"id":5,"title":6,"authors":7,"badge":10,"body":11,"date":504,"description":505,"extension":506,"image":507,"meta":508,"navigation":509,"path":510,"seo":511,"stem":512,"__hash__":513},"blog/blog/how-many-languages-do-you-support.md","\"How many languages do you support?\" — and why our honest answer is six numbers, not one",[8],{"name":9},"The Mind.com Team","Methodology",{"type":12,"value":13,"toc":490},"minimark",[14,18,22,28,36,39,44,136,153,156,159,161,165,168,190,193,196,198,202,213,218,225,227,231,251,260,263,266,291,294,296,300,303,310,321,332,335,347,349,353,363,366,377,380,382,386,389,420,423,425,429,449,451,455,484,487],[15,16,6],"h1",{"id":17},"how-many-languages-do-you-support-and-why-our-honest-answer-is-six-numbers-not-one",[19,20,21],"p",{},"It is the first question on every procurement call and the first question in every demo. We get it three times a week:",[23,24,25],"blockquote",{},[19,26,27],{},"\"How many languages does InterMIND support?\"",[19,29,30,31,35],{},"The honest answer is: ",[32,33,34],"strong",{},"it depends on which surface you mean."," We have six of them, and they have different language lists for good reasons. This page is the definitive answer so we stop giving inconsistent ones.",[37,38],"hr",{},[40,41,43],"h2",{"id":42},"the-short-version","The short version",[45,46,47,61],"table",{},[48,49,50],"thead",{},[51,52,53,57],"tr",{},[54,55,56],"th",{},"Surface",[54,58,60],{"align":59},"right","Languages",[62,63,64,80,90,99,108,118],"tbody",{},[51,65,66,75],{},[67,68,69,70,74],"td",{},"Platform API (",[71,72,73],"code",{},"createParticipant"," accepts)",[67,76,77],{"align":59},[32,78,79],{},"24",[51,81,82,85],{},[67,83,84],{},"Real-time voice translation in meetings",[67,86,87],{"align":59},[32,88,89],{},"21",[51,91,92,95],{},[67,93,94],{},"Real-time chat-message translation",[67,96,97],{"align":59},[32,98,89],{},[51,100,101,104],{},[67,102,103],{},"Real-time shared-notes translation",[67,105,106],{"align":59},[32,107,89],{},[51,109,110,113],{},[67,111,112],{},"On-demand file translation in chat (DeepL Document API)",[67,114,115],{"align":59},[32,116,117],{},"30",[51,119,120,131],{},[67,121,122,123,130],{},"Website UI (localized at ",[124,125,129],"a",{"href":126,"rel":127},"https://intermind.com",[128],"nofollow","intermind.com",")",[67,132,133],{"align":59},[32,134,135],{},"6",[23,137,138],{},[19,139,140,143,144,146,147,149,150,152],{},[32,141,142],{},"The product is 21 languages."," Files: ",[32,145,117],{},". Site: ",[32,148,135],{},". API engine: ",[32,151,79],{},".",[19,154,155],{},"Voice, chat, and notes share the same 21. Files are wider because the file pipeline can be. The website is narrower because translating a marketing site is a different cost than translating in-product copy.",[19,157,158],{},"Everything below is why those numbers are not the same.",[37,160],{},[40,162,164],{"id":163},"why-there-isnt-a-single-number","Why there isn't a single number",[19,166,167],{},"A translation platform is not one product. It is at least three:",[169,170,171,178,184],"ol",{},[172,173,174,177],"li",{},[32,175,176],{},"A real-time speech pipeline"," — audio in, translated audio out under one second, in a meeting where the latency budget is brutal.",[172,179,180,183],{},[32,181,182],{},"A real-time text pipeline"," — chat messages and shared notes, character-by-character, propagated to every viewer in their language.",[172,185,186,189],{},[32,187,188],{},"An asynchronous document pipeline"," — a 40-page PDF dropped into chat, translated as a whole file with formatting and structure preserved.",[19,191,192],{},"Each pipeline has its own engine, its own latency budget, and its own per-pair quality envelope. A language that is production-grade for documents (slow, careful, many passes) can be unusable for live voice (fast, one shot, no retries). Treating those as \"one language count\" is what produces marketing pages that say \"200+ languages\" and tell you nothing.",[19,194,195],{},"So we publish per-surface.",[37,197],{},[40,199,201],{"id":200},"_1-platform-api-24-languages","1. Platform API — 24 languages",[19,203,204,205,212],{},"This is the widest list, because it is the engine surface. The ",[124,206,209,211],{"href":207,"rel":208},"https://api.mind.com/#operation/createParticipant",[128],[71,210,73],{}," endpoint"," accepts these 24 codes:",[19,214,215],{},[71,216,217],{},"ar, cs, da, de, en, es, fi, fr, hi, hu, is, it, ja, ko, nl, no, pl, pt, ro, ru, sv, tr, uk, zh",[19,219,220,221,224],{},"If you are an integrator building on Mind directly, you can hand any of those 24 to a participant. That ",[32,222,223],{},"does not"," mean every one of those 24 meets our product quality bar. It means our engine can emit text in that language. The product gates this further.",[37,226],{},[40,228,230],{"id":229},"_2-voice-chat-and-notes-inside-the-product-21-languages","2. Voice, chat, and notes inside the product — 21 languages",[19,232,233,234,236,237],{},"When a user joins an InterMIND meeting, the language picker shows ",[32,235,89],{}," languages, not 24. Three are removed: ",[32,238,239,240,243,244,247,248,250],{},"Arabic (",[71,241,242],{},"ar","), Hindi (",[71,245,246],{},"hi","), Turkish (",[71,249,51],{},").",[19,252,253,254,259],{},"Why those three: each of those pairs does not clear the median-quality bar we publish on ",[124,255,257],{"href":256},"/benchmark",[71,258,256],{},". The engine produces output. The output is not good enough to put in front of a regulated meeting. So we hide them from the picker rather than letting a user pick a language that will embarrass them in front of an auditor.",[19,261,262],{},"This is a deliberate choice. Most of the category goes the other way — list everything the model can emit, let the buyer find out at the wrong moment that one of those pairs is broken. We would rather show a smaller picker.",[19,264,265],{},"The same 21-language list applies across the three real-time surfaces:",[267,268,269,275,285],"ul",{},[172,270,271,274],{},[32,272,273],{},"Voice translation"," — per-viewer translated audio with sub-second latency. Each participant picks their own listening language at the start of the call.",[172,276,277,280,281,250],{},[32,278,279],{},"Chat messages"," — every message translated as it is typed; edits produce per-language diffs (",[124,282,284],{"href":283},"/blog/intermind-v1-2-release","see v1.2",[172,286,287,290],{},[32,288,289],{},"Shared notes"," — character-by-character live translation of the host's notes pane, per viewer, with diff history.",[19,292,293],{},"One picker, one set of 21 languages, three places it shows up.",[37,295],{},[40,297,299],{"id":298},"_3-on-demand-file-translation-30-languages","3. On-demand file translation — 30 languages",[19,301,302],{},"Drop a PDF, DOCX, PPTX, or XLSX into the chat. Each participant can request the file in their language. The translated copy is returned as the same file format with structure preserved (tables, headings, lists).",[19,304,305,306,309],{},"This surface uses the ",[32,307,308],{},"DeepL Document API",", which currently supports 30 target languages — wider than our real-time voice pipeline. We pass that capability straight through. If you can translate the PDF to Estonian on DeepL today, you can translate it to Estonian in InterMIND chat today.",[19,311,312,313,316,317,320],{},"The DeepL list includes some languages our real-time pipeline does not — for example ",[32,314,315],{},"Bulgarian, Greek, Estonian, Indonesian, Lithuanian, Latvian, Norwegian Bokmål, Slovak, Slovenian"," — and, notably, two of the three we filter from voice: ",[32,318,319],{},"Arabic and Turkish work for files."," A French participant can request the contract PDF in Turkish even though they cannot listen to the meeting in Turkish. We flag this asymmetry in the picker rather than hide it.",[19,322,323,324,327,328,331],{},"The one language in our API that is ",[32,325,326],{},"not"," available on either pipeline: ",[32,329,330],{},"Hindi."," The engine accepts it as an input, but neither voice quality nor DeepL coverage gets us to a shippable result yet. It is the single language that is \"on the API, off everywhere else.\"",[19,333,334],{},"Why the file list is bigger in general:",[267,336,337,344],{},[172,338,339,340,343],{},"Documents are ",[32,341,342],{},"asynchronous",". There is no one-second budget, so the pipeline can afford a slower, more careful engine that handles more pairs well.",[172,345,346],{},"DeepL is the best-in-class document-translation engine and we use it directly for files. We do not try to route documents through the same engine that runs voice. Different problem, different tool.",[37,348],{},[40,350,352],{"id":351},"_4-website-ui-6-languages","4. Website UI — 6 languages",[19,354,355,358,359,362],{},[124,356,129],{"href":126,"rel":357},[128]," is currently shipped in ",[32,360,361],{},"6 locales",": English, German, Spanish (Latin America), French, Portuguese (Brazilian), and Chinese (Simplified).",[19,364,365],{},"Why much narrower than the product:",[267,367,368,371,374],{},[172,369,370],{},"A marketing site has its own translation cost — and its own quality bar. Landing pages, FAQs, pricing, legal copy. It is not free.",[172,372,373],{},"We ship locales where traffic and pipeline justify the maintenance, not for every language the product supports.",[172,375,376],{},"The 6 chosen are the ones our actual organic and procurement traffic comes from this quarter.",[19,378,379],{},"This list will grow as pipeline grows. It is not capped at six on principle. It is capped at six on cost.",[37,381],{},[40,383,385],{"id":384},"the-pattern-and-how-to-read-it-on-any-vendor-page","The pattern, and how to read it on any vendor page",[19,387,388],{},"Every multilingual vendor has these same surfaces. Most of them quote you the widest one and let you discover the others on your own. The honest version is:",[267,390,391,398,404,414],{},[172,392,393,394,397],{},"The ",[32,395,396],{},"engine"," list is widest, because it is what the model emits.",[172,399,393,400,403],{},[32,401,402],{},"product"," list is narrower, because not every engine output meets a product bar.",[172,405,393,406,409,410,413],{},[32,407,408],{},"document"," list often differs from the ",[32,411,412],{},"voice"," list, because the engines are different.",[172,415,393,416,419],{},[32,417,418],{},"website"," list is narrowest, because translating your own site costs you cash.",[19,421,422],{},"When you evaluate any platform in this category, ask for the four numbers separately. If a vendor refuses to split them — or doesn't know — you are not getting a useful answer.",[37,424],{},[40,426,428],{"id":427},"what-we-will-not-pretend","What we will not pretend",[267,430,431,437,443],{},[172,432,433,436],{},[32,434,435],{},"The three filtered languages are real."," Our engine accepts Arabic, Hindi, and Turkish. We pull them from the voice/chat/notes picker because they don't clear our public quality bar. Arabic and Turkish are still available on the file pipeline (DeepL handles those documents well). Hindi is unavailable on either real-time or file — we won't ship it until either side gets there.",[172,438,439,442],{},[32,440,441],{},"The file list (30) being wider than the voice list (21) is a real gap, not a feature."," A French user can request an Estonian translation of the PDF but cannot listen to the meeting in Estonian. We will not paper over this by quoting the bigger number and hoping you don't notice.",[172,444,445,448],{},[32,446,447],{},"The website-locale list (6) is product debt, not product principle."," We ship more as traffic justifies it.",[37,450],{},[40,452,454],{"id":453},"try-it-yourself","Try it yourself",[267,456,457,469,476],{},[172,458,459,464,465,152],{},[124,460,462],{"href":461},"/demo",[71,463,461],{}," — runs the production voice + chat pipeline against your audio, in any of the 21 product languages. The same pipeline that scores ",[124,466,467],{"href":256},[71,468,256],{},[172,470,471,475],{},[124,472,473],{"href":256},[71,474,256],{}," — per-pair, per-month quality on real traffic. Includes the pairs we deliberately hide from the picker, deep-linkable by URL.",[172,477,478,483],{},[124,479,481],{"href":480},"/benchmark/methodology",[71,482,480],{}," — what the numbers are, what they aren't, who the judge is.",[19,485,486],{},"Six numbers, six surfaces, one engine. That is the honest answer to \"how many languages do you support.\"",[19,488,489],{},"— The Mind.com Team",{"title":491,"searchDepth":492,"depth":493,"links":494},"",2,3,[495,496,497,498,499,500,501,502,503],{"id":42,"depth":492,"text":43},{"id":163,"depth":492,"text":164},{"id":200,"depth":492,"text":201},{"id":229,"depth":492,"text":230},{"id":298,"depth":492,"text":299},{"id":351,"depth":492,"text":352},{"id":384,"depth":492,"text":385},{"id":427,"depth":492,"text":428},{"id":453,"depth":492,"text":454},"2026-05-24","Every vendor quotes one language count. We can't, because translation isn't one product. Here is the per-surface breakdown for InterMIND — what is filtered, why, and what we publish on the website.","md","/blog/how-many-languages-do-you-support.svg",{},true,"/blog/how-many-languages-do-you-support",{"title":6,"description":505},"blog/how-many-languages-do-you-support","WTxf6eR1MzAuUhfssvJ_6Tm34kY9hANGOzeAnvJNCbY",[515,520],{"title":516,"path":517,"stem":518,"description":519,"children":-1},"The false-fluency trap: when \"good-enough\" English is worse than no English","/blog/false-fluency-trap","blog/false-fluency-trap","Aviation regulators mandated it after Tenerife and Avianca. Medical literature quantifies it down to error counts per patient encounter. Forty years of international-business research documents the same mechanism in the boardroom. Why partial fluency in a shared language is worse than no shared language at all — with the data.",{"title":521,"path":522,"stem":523,"description":524,"children":-1},"Inside the four translation pipelines that run InterMIND","/blog/inside-the-translation-pipelines","blog/inside-the-translation-pipelines","There is no \"the translation\" in InterMIND. There are four pipelines — voice, chat, notes, documents — each with its own engine, latency budget, and quality envelope. This is what actually happens between the moment you speak and the moment a participant in another language understands you."]